Northeastern Section - 48th Annual Meeting (18–20 March 2013)

Paper No. 9
Presentation Time: 4:10 PM

DISPLAYING GEOREFERENCED INTERACTIVE VIRTUAL SPECIMENS IN GOOGLE EARTH USING AUTODESK 123D CATCH


DE PAOR, Declan G., Physics Department, Old Dominion University, 235 OceanographyandPhysics Bldg, Norfolk, VA 23529, ddepaor@odu.edu

Collection of specimens for identification and analysis is an important part of field-based geoscience research and education. Extensive specimen collections are common in long-established departments and museums but rare in community colleges. In distance education, cumbersome postal or visiting arrangements are necessary to get specimens into the hands of students, and even for traditional students, access to specimens is limited before and after scheduled lab times. While virtual specimens may never fully match physical specimens, they offer great opportunities for augmenting reality in the field component of courses delivered online. They can also be important research objects, for example as archives of specimens that are destroyed by analysis or as models that can be instantly shared among experts across the globe.

We have created virtual specimens and georeferenced them on Google Earth using a variety of methods. We first scanned hand samples of rocks, minerals, crystals, and fossils using a NextEngine 3D laser scanner, converted them to COLLADA models with the Digital Asset Exchange format, and imported them into Google Earth using both the desktop application’s model function and the Google Earth web browser plug-in and API. In the latter case, specimens are manipulatable with Javascript controls.

This approach works well but is limited by hardware and time requirements. We developed an alternative technique using camera images taken from multiple directions and assembled using Google (now Trimble) SketchUp. This was low-cost but was cumbersome for all but the simplest of specimen geometries. However a new development has radically improved the photographic method. Using Autodesk 123D Catch, a virtual specimen can be made with ease, based on 20 - 40 photographs. Editing can be done within Catch or in a third party modeling program such as MeshLab. After conversion from 3dp to dae file format, models can be georeferenced into the Google Earth terrain.

An advantage of virtual specimens is that they can be any size. Whole outcrops or landscape segments can be treated as if they were hand specimens. Given the ease with which non-technical users can now create virtual specimens there is an opportunity for crowd-sourcing to create a vast, worldwide, georeferenced geological specimen collection.